Vol.57'] X7PPER GREENSAND AND CHLORITIC MARL OF WILTSHIRE. 123 



the former contains a fauna comparable with that of the Cornstones 

 and the Rye Hill Sand ^ ; while the base of the latter is a glauconitic 

 chalk which has a Chalk-Marl fauna, but does not contain Stauro- 

 nema. Here, therefore, the sub-zone of Stauronema appears to be 

 absent; and a fauna resembling that of Rye Hill and Maiden 

 Bradley occurs in a bed which is inseparable from the Greensand, 

 and strongly marked off from the Chalk. The same is the case 

 both in Western and in Southern Dorset, many exposures occurring 

 in both districts and displaying the same general features. 



In the Isle of Wight the circumstances are different: the 

 Stauronema-XiQ&s are well developed, and in most places have a 

 nodule-bed at their base, which rests on a piped surface of laminated 

 sand. This sand, however, does not contain many fossils, and, if 

 comparable vsdth anything at Maiden Bradley, it may be paralleled 

 with the sand below the Cornstone-Bed. Here, therefore, on the 

 contrary, we have no satisfactory base to the Lower Chalk, without 

 including something that may correspond to the Rye Hill Sand. 



In Northern Wiltshire, near Urchfont and Devizes, there is again 

 a well-marked base to the Stauronema-zone^ and an absence of any 

 beds exactly comparable to the upper part of the Rye Hill Sand. 



The evidence of these other localities is therefore conflicting, 

 and it cannot be denied that the systematic geologist is here 

 confronted with the horns of a dilemma : he must either exclude 

 the Rye Hill fauna from the zone of Ammonites varians, although 

 the cephalopoda of the Chalk-Marl are prominent members of it ; 

 or else he must include in that zone beds which are in some places 

 separated by a strong physical and ontological break. 



If the succession in Dorset and Northern Wiltshire were like that 

 in Southern Wiltshire, we should have no hesitation in so dividing the 

 two formations as to place the Rye Hill fauna in the Cenomanian or 

 Lower Chalk stage, because that would seem a natural inference 

 from the evidence which we have recorded in this paper. We feel, 

 moreover, that such an arrangement would bring the Cenomanian 

 of England more into Hne with the Cenomanian of French geologists, 

 who insist upon including within it every bed that contains 

 Ammonites varians. 



It is consequently with some reluctance that we retain the 

 arrangement that is at present accepted, and we wish it to be clearly 

 understood that we do so only for the sake of convenience, and in 

 spite of the evidence adduced. in the foregoing pages. It is, in 

 fact, one of those cases in which the palseontological is in conflict 

 with the stratigraphical evidence. If the break in Dorset were not 

 where we believe it to be, we could follow the lead of palaeontology ; 

 but, in the present state of our knowledge, we prefer to take the 

 stratigraphical line in Dorset as the least awkward of the two 

 alternatives. 



^ See Proc. Dorset Nat. Hist. & Ant. Soc. vol. xvii (1896) p. 99, & Mem. 

 Geol. Surv. ' Cretaceous Kocks of Britain' vol. i (1900) pp. 243 et seqq. 



